paul-corbalan
/
Scan-Statistics-Project-4Y-INSA
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Experience plan work with ScoreDistribEmpiric

This commit is contained in:
Paul-Corbalan 2022-04-12 17:35:16 +02:00
parent 2231672fb0
commit 7193c45e5a
1 changed files with 27 additions and 52 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

77
Comparaison_of_methods.rmd
View File

@ -211,37 +211,6 @@ ScS_H1
## 3. Local score
### Distribution of scores via Monte Carlo
```{r}
# Calcul du choix de E
E = 1
maxXk = floor(E*(log(lambda1/lambda0)))
while (maxXk < 3) {
E = E+1
maxXk = floor(E*(log(lambda1/lambda0)))
}
ppH0 = PoissonProcess(lambda0,10^4)
n1 = length(ppH0)
tbe0 = ppH0[2:n1]-ppH0[1:n1-1]
print(ks.test(tbe0, 'exp'))
xp = floor(E*(log(lambda1/lambda0)+(lambda0-lambda1)*tbe0)) # ne pas mettre le floor ni le E (certes égale à 1)
min_X = min(xp)
max_X = max(xp)
vect.score = min_X:max_X
P_X = table(factor(xp, levels = min(xp):max(xp)))
P_X = P_X/sum(table(xp))
Mean_xp = sum(vect.score*P_X)
print(Mean_xp)
#print(dist.theo.scores) # Mettre à jour avec Elisa
#print(P_X)
```
```{r}
ComputeE <- function(lambda0, lambda1){
E = 1
@ -256,28 +225,26 @@ ComputeE <- function(lambda0, lambda1){
```
```{r}
ScoreDistrib <- function(lambda0, lambda1, NbSeq, T){
ScoreDistribEmpiric <- function(lambda0, lambda1, n_sample, T){
E = ComputeE(lambda0, lambda1)
Score=c()
for (i in 1:NbSeq) {
selected
for (i in 1:n_sample){
ppH0 = PoissonProcess(lambda0,T)
n1 = length(ppH0)
tbe0 = ppH0[2:n1]-ppH0[1:n1-1]
X = floor(E*(log(lambda1/lambda0)+(lambda0-lambda1)*tbe0))
Score=c(Score,X)
}
ppH0 = PoissonProcess(lambda0,T)
n1 = length(ppH0)
tbe0 = ppH0[2:n1]-ppH0[1:n1-1]
# print(ks.test(tbe0, 'exp'))
min_X = min(Score)
max_X = max(Score)
X = floor(E*(log(lambda1/lambda0)+(lambda0-lambda1)*tbe0)) # ne pas mettre le floor ni le E (certes égale à 1)
P_X = table(factor(Score, levels = min_X:max_X))/sum(table(Score)).Freq
df = data.frame("Score_X" = min(Score):max(Score), "P_X" = P_X)
df <- df[,-2]
min_X = min(X)
max_X = max(X)
vect.score = min_X:max_X
P_X = table(factor(X, levels = min_X:max_X))
P_X = P_X/sum(table(X))
return (list("X" = X, "P_X" = P_X))
return (df)
}
```
```{r}
@ -321,7 +288,6 @@ LocaScoreMC <- function(lambda0, lambda1, NbSeq, T, X_seq, P_X, tbe0){
pvalue = c(pvalue, daudin_result)
}
print(NbSeqH0)
LS_H0=data.frame(num=1:NbSeq, pvalue_scan=pvalue, class=(pvalue<0.05))
return(LS_H0)
}
@ -331,7 +297,7 @@ LocaScoreMC <- function(lambda0, lambda1, NbSeq, T, X_seq, P_X, tbe0){
```{r}
NbSeq = 10**3
T = 10
for (lambda0 in (1:5)){
for (lambda0 in (2:5)){
for (lambda1 in c(2,4,6)){
if (lambda0 < lambda1){
cat("Nb = ", NbSeq, ", lambda0 = ", lambda0, ", lambda1 = ", lambda1, "\n", sep = "")
@ -344,7 +310,16 @@ for (lambda0 in (1:5)){
tbe0[[i]]=tbei
}
Score = ScoreDistrib(lambda0, lambda1, NbSeq, T)
Score = ScoreDistribEmpiric(lambda0, lambda1, NbSeq, T)
X_seq = Score$Score_X
P_X = Score$P_X
LS_H0 = LocaScoreMC(lambda0, lambda1, NbSeq, T, X_seq, P_X, tbe0)
print(summary(LS_H0))
cat("-\n")
Score = ScoreDistribElisa(lambda0, lambda1, NbSeq, T)
X_seq = Score$X
P_X = Score$P_X